Magazines/Newspapers

Abrupt Decrease in Tropical Pacific Sea Surface Salinity at End of Little Ice Age

Description
Abrupt Decrease in Tropical Pacific Sea Surface Salinity at End of Little Ice Age
Published
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  mental growth as described in ( 21 ). Resolution de-creased with ontogeny because of diminished incre-ment width with age.   18 3    18 O otolith  ratios arereported relative to Pee Dee belemnite (PDB) stan-dard, and water oxygen isotope ratios are reportedrelative to Standard Mean Ocean Water (SMOW)using conventional per mil (‰) notation. Precision of isotope analyses is as follows: conventional H 3 PO 4 extraction (of aragonite) is  0.1‰ (1  ), laser abla-tion extraction (of aragonite) is   0.2‰ (1  ), andCO 2  equilibration (of water) is    0.1‰ (1  ). Preci-sion was calculated based on between- and within-run comparisons of working standards.22. SST data from the Joint Institute for the Study of the Atmosphere and the Oceans database can befound online at http://tao.atmos.washington.edu/data_sets/for Puerto Chicama (8°S, 1989–1999)and Paita (5°S, 1989–1998), and measured  18 O water   values (–0.5    0.2‰ 1  ;  n    4) wereapplied to the aragonite   18 O temperature equa-tion ( 24 ) to build models of    18 O otolith  profiles.23. Figure 3A compares model   18 O otolith  values to thecomposite mean  18 O of 10 modern otoliths. The com-posite was constructed by calculating the mean   18 Ovalue of each increment in every otolith. These valueswere averaged with all other contemporaneous incre-ments to produce the 6-month time-average compos-ite. The mean  18 O of the modern otoliths is –0.1‰(1  0.4), with a mean seasonal  18 O range of 1‰(1    0.3). Modeled and measured isotope profileswere strongly correlated, thus documenting  18 O otolith records as valid proxies for SSTs.24. E. L. Grossman, T. L. Ku,  Chem .  Geol .  59 , 59 (1986).25. S. Pozorski, T. Pozorski,  Ann .  Carnegie Mus .  48 , 337(1979).26.    ,  J .  Field Archaeol .  13 , 381 (1986).27. E. J. Reitz, unpublished data.28. C. E. Shannon, W. Weaver,  The Mathematical Theory of Communication  (Univ. of Illinois Press, Urbana, IL,1949).29. S. H. Hurlbert,  Ecology   52 , 577 (1971).30. D. Pauly, V. Christensen,  Nature  374 , 255 (1995).31. R. Froese, D. Pauly,  Fishbase , online at www.fishbase.org (2 March, 2001).32. Diversity was calculated by  H    S  i   1 (p i  )(log  p i  ),where  H  diversity,  S  i   1  the sum of all taxa, and  p i   relative abundance of   i  th taxon ( 28 ). Equitabilitywas calculated by  V     H  /log  S , where  V     equi-tability and  S    number of taxa ( 29 ). Trophic leveldetermination was based on data in ( 31 ), or thenearest approximation based on analog species. Tro-phic value was determined for each taxon identifiedand multiplied by its MNI, and then the total of alltrophic values was divided by total-site MNI. All siteswere excavated in a similar fashion, and we used finemesh-screen recovery, thus minimizing recovery biasand ensuring representative samples.33. D. H. Sandweiss, K. A. Maasch, D. F. Belknap, J. B.Richardson III, H. B. Rollins,  J .  Coastal Res .  14 , 367(1998).34. C. Perrier, C. Hillaire-Marcel, L. Ortlieb,  Geogr  .  Phys . Quat  .  48 , 23 (1994).35. We thank B. McClain for assistance with isotopicanalysis and M. Cornejo for field assistance. Funded inpart by grants from the Geological Society of Amer-ica (C.F.T.A.), Explorer’s Club International (C.F.T.A.),NSF grant ATM-0082213 (D.E.C.), The Heinz Charita-ble Trust (D.H.S.), the University of Maine FacultyResearch Fund (D.H.S.), and Department of Energygrant DE-FC09-96SR18546 (C.S.R.).26 April 2001; accepted 14 January 2002 Abrupt Decrease in TropicalPacific Sea Surface Salinity atEnd of Little Ice Age Erica J. Hendy, 1 * Michael K. Gagan, 1 Chantal A. Alibert, 1 Malcolm T. McCulloch, 1  Janice M. Lough, 2 Peter J. Isdale 2 A420-yearhistoryofstrontium/calcium,uranium/calcium,andoxygenisotoperatios in eight coral cores from the Great Barrier Reef, Australia, indicates thatsea surface temperature and salinity were higher in the 18th century than inthe 20th century. An abrupt freshening after 1870 occurred simultaneouslythroughout the southwestern Pacific, coinciding with cooling tropical temper-atures. Higher salinities between 1565 and 1870 are best explained by acombination of advection and wind-induced evaporation resulting from astrong latitudinal temperature gradient and intensified circulation. The globalLittle Ice Age glacial expansion may have been driven, in part, by greater poleward transport of water vapor from the tropical Pacific. The Little Ice Age (LIA) appears in most Northern Hemisphere paleoclimate recon-structions as multiple, century-scale periodsof anomalously cold, dry conditions betweenthe 15th and late 19th centuries ( 1–4 ). Gla-cial advances in both hemispheres ( 1 ) and enhanced polar atmospheric circulation ( 5 )suggest that the LIA was a global-scale event.With the exception of the Quelccaya ice corerecord from equatorial Peru ( 6  ), however, thesparse reconstructions of the LIA availablefrom the tropics and Southern Hemispherefail to identify synchronous cold periods ( 7,8 ). An alternative scenario is that coolingduring the LIA was restricted to higher lati-tudes ( 9, 10 ). We examine the nature of theLIA in the tropical southwestern Pacific since1565 using coral proxies to reconstruct seasurface temperature (SST) and salinity(SSS), the key indicators of climate changewithin the tropical ocean-atmosphere system.We present 420-year records of three cor-al paleoclimate tracers,   18 O, Sr/Ca, and U/Ca, constructed from replicated measure-ments of eight cores from massive  Porites  sp.colonies (Fig. 1). The cores were collected atseven reefs, 12 to 120 km apart, from thecentral Great Barrier Reef (GBR), Australia[for details and site maps, see ( 11 )]. The useof multiple cores allows us to test the fidelityof individual tracers over century time scalesand to establish regional-scale proxy climatesignals. Coral  18 O, the most frequently used coral climate proxy tool, reflects a combina-tion of SST and the seawater   18 O composi-tion. The latter responds to changes in SSScaused by shifts in seawater    18 O produced  by evaporation and freshwater input. By us-ing parallel measurements of the coral paleo-thermometers Sr/Ca and U/Ca to determineSST, we are able to separate SST variationfrom changes in seawater   18 O ( 12, 13 ). Thisapproach allows us to resolve both SST and SSS over the past four centuries.Five-year bulk increments were sampled for equivalent periods across all eight coresafter cross-dating with ultraviolet (UV) flu-orescent bands; x-radiography was used toreveal annual density banding ( 14 ). Fine powder was milled from 2-mm square-sec-tion grooves along the center of the coralgrowth axis, homogenized, and subsampled for stable isotope and trace element analysis( 15 ). Figure 1 shows the replicated measure-ments combined into a composite record,with the variability between corals shown asa 95% confidence envelope calculated for each pentannual interval ( 16  ). Individual  18 O records reproduce not only the century-scale trends in the composite   18 O recon-struction but also decadal variations of up to0.26‰, capturing on average 72% of theshared    18 O signal. Intercoral variability isgreater between individual Sr/Ca and U/Carecords, which share on average 33% and 26%, respectively, of the common variance.When converted to temperature by applyingaccepted SST-slope calibrations ( 17, 18 ), theSr/Ca record is better constrained than theU/Ca record (Sr/Ca,  0.3°C; U/Ca,  0.5°C).This intercoral Sr/Ca variability is equivalentto that found between high-resolution season-al records from  Porites  colonies at sites closeto those used here ( 13, 17  ). However, multi- ple Sr/Ca or U/Ca records are required toreconstruct decadal to century-scale SST sig-nals, which in the GBR are one-tenth the sizeof seasonal variations.The Sr/Ca and U/Ca composite records,converted to SST anomalies (SSTA), are ver-ified against GISST2.2 ( 19 ) for the 1° grid at18°S, 146°E (resampled to 5-year averages,1905 to 1985). Linear regressions are signif- 1 Research School of Earth Sciences, Australian Na-tional University, Canberra, ACT 0200, Australia. 2 Australian Institute of Marine Science, PMB 3,Townsville M.C., Queensland 4810, Australia.*To whom correspondence should be addressed. E-mail: erica.hendy@anu.edu.au R  E P O R T S www.sciencemag.org SCIENCE VOL 295 22 FEBRUARY 2002  1511  icant at the 99% level for both Sr/Ca ( r    0.68) and U/Ca ( r     0.63). The Sr/Ca and U/Ca SSTA reconstructions are in excellentagreement ( r     0.72,  P     0.001,  n    81).Although the temperature calibrations ( 17,18 ) were developed from short high-resolu-tion records, both reconstructions successful-ly reproduce the 0.7°C warming evident over the 80-year GISST2.2 series (Sr/Ca,  0.7°C;U/Ca,  1.0°C). Furthermore, the coral paleo-temperature reconstructions compare reason-ably well with the U.K. Meteorological Of-fice tropical western Pacific SSTA compositerecord for 1865 to 1985 [20°N to 20°S, 120°Eto 170°W ( 20 )], despite significant uncertain-ty in the instrumental record before   1900(Sr/Ca,  r   0.58; U/Ca,  r   0.58;  P   0.005)(Fig. 2, A to C). This indicates that the coralSSTA reconstructions from the GBR are re-gionally significant.The Sr/Ca and U/Ca records reconstructSSTs that were   0.2° to 0.3°C cooler thanthe long-term average during an extended  period from 1565 to 1700 (Fig. 2, A and B).After a century-scale warming of 0.4°C cen-tered on 1700, above-average SSTs persistthrough most of the 18th and 19th centuries before cooling to a minimum in the early 20thcentury. From this cold interval, the SSTAreconstructions capture the 20th centurywarming until the 1980s, when the coralcores were collected. It is conspicuous thatthe period from the 1700s to the 1870s wasconsistently as warm as the early 1980s. Theonly other Pacific coral Sr/Ca record, fromRarotonga (Fig. 2D) ( 21 ), also reconstructsSSTs for the 18th and 19th centuries that areas warm as, or warmer than, the 20th century.The question raised by our coral SSTA re-constructions is whether the positive tempera-ture anomalies during the LIA were limited tothe tropical Pacific (or to the tropics as a whole)or penetrated throughout the Southern Hemi-sphere. The small number of proxy temperaturerecords from the Southern Hemisphere, and their conflicting signals, limit attempts to de-scribe regional patterns ( 4, 7, 8 ) (Fig. 2, E and F). In contrast, global proxy temperature com- posites consistently show that the 20th centurywas significantly warmer than the four preced-ingcenturies( 7,22–24 ).Availableproxysourc-es, however, are predominantly land-based and from the temperate Northern Hemisphere, and therefore are unlikely to be representative of theglobal-scale response. The recovery to warmer conditions seen in the global and hemisphericcomposites coincides with the glacial retreatthat signaled the end of the LIA in the late 19thto early 20th centuries ( 1 ), while at the sametime, the tropical Pacific cooled according toour coral SSTA reconstructions. If the LIAcooling was limited to higher latitudes, thenlatitudinal SST gradients were greater than at present, in turn intensifying the poleward trans- port of heat in the atmosphere and ocean ( 9,10 ). Evaporation and condensation are effective processes for redistributing atmospheric heat between the tropics and extratropics. Measure-ments of coral   18 O enable us to monitor changes in the tropical hydrological balancethrough the isotopic change in seawater as it isenriched in  18 O by evaporation and depleted in 18 O by freshwater input ( 25 ).The most striking feature of the composite420-year   18 O GBR coral record is the abrupt Fig. 1.  GBR compositerecords of ( A ) Sr/Ca,( B ) U/Ca, and ( C )  18 Oat pentannual resolu-tion. Solid lines showthe average recon-struction ( 16 ) normal-ized to the period1860 to 1985, plottedas ratios (left axis) andSSTA (right axis). The95% confidence inter-vals are shown by thedotted lines surround-ing the solid line of the reconstruction.SSTA conversions are–61.5   mol/mol per °C for Sr/Ca ( 17 ) and–46.5 nmol/mol per °C for U/Ca ( 18 ). Thehorizontal dashedlines define a 1°Cband according tothese slope calibra-tions. ( D ) The number of records averaged ateach pentannual in-terval for the  18 O re-construction. Fig. 2.  A comparison of coraland instrumental temperaturerecords. GBR composite SSTAreconstructions were derivedfrom coral Sr/Ca ( A ) and U/Ca( B ) and were converted toSSTA using slope calibrationsfrom ( 17 ) and ( 18 ), respective-ly. ( C ) UKMO composite SSTArecords for the tropical west-ern Pacific [20°N to 20°S,120°E to 170°W ( 20 )]. ( D ) TheRarotonga Sr/Ca SST record(21.5°S, 159.5°W) of Linsley  et al.  ( 21 ) is plotted using theSr/Ca SST slope calibration giv-en in ( 21 ). ( E  and  F ) Proxy tem-perature (land and SST) com-pilations for the Southern andNorthern Hemispheres, respec-tively ( 7 ). Jones  et al.  ( 7 ) cau-tioned that the SouthernHemisphere reconstruction (E)is a poor representative of temperature because it is con-structed from only seven pa-leotemperature records. Allrecords plotted for comparisonare resampled to equivalent5-year averages, and all seriesare normalized to the commonperiod 1860 to 1985. R  E P O R T S 22 FEBRUARY 2002 VOL 295 SCIENCE www.sciencemag.org 1512  0.2‰ shift between 1850 and 1870 toward lower modern  18 O values (Fig. 3A). Superim- posed on this century-scale signal are interdec-adal  18 O cycles of up to 0.26‰. Only a veryweak relationship is seen between the coral  18 O and paleotemperature proxies. By takinginto account the SST anomalies recorded by thecoral Sr/Ca, we are able to subtract the compo-nent of    18 O related to temperature and toresolvethesalinityeffectfromtheresidual( 26  ).The residual   18 O (Fig. 3B) shares more than80% of its variance with the srcinal   18 Oreconstruction, indicating that the long-term 18 O depletion since the 1870s was the result of a significant freshening of the GBR lagoonwaters in the modern period. This late-19thcentury shift is recorded synchronously by  18 O coral records throughout the Coral Sea(Fig. 3), including Abraham Reef on the south-ernouteredgeoftheGBR( 27  ),NewCaledonia( 28 ), and Espirtu Santos, Vanuatu ( 29 ). Sup- porting a southwestern Pacific–wide SSS-dom-inated    18 O signal is the lack of any relation-ship between instrumental SST and either theVanuatu or Abraham Reef   18 O records ( 30 ). Ithas also been proposed that salinity is moreinfluential at frequencies lower than interannual periods in the New Caledonia coral  18 O signal( 31 ). These results suggest that, in addition towarm SSTs, conditions in the tropical south-western Pacific during the LIA were also con-sistently more saline.Reduced precipitation during the LIA, asso-ciated with a change in the position of theIntertropical Convergence Zone, was suggested to explain a similar century-scale   18 O shiftaround 1850 seen in the Gulf of Chiriquı´  18 Ocoral record in the Eastern Pacific ( 32 ). Thestrength of the Australian summer monsooninfluences SSS in the GBR ( 33 ) and is animportant source of the interdecadal coral  18 Ovariability [ r   0.74 for   18 O and Queensland summer rainfall index 1890– 1985 ( 34 )]. How-ever, as explained below, three centuries of drier conditions cannot alone account for the  18 O enrichment in the GBR corals before1870. Twenty-six percent of the land area of Queensland drains into the GBR, and dischargerecords from the largest freshwater contributor,the Burdekin River, also correlate strongly withcoral  18 O [ r   0.70, 1895–1985 ( 35 )]. Extrap-olation of this relationship would imply that theriverceasedflowingformostoftheperiodfrom1565 to 1860. Such an extreme hydrologicalreconstruction is contradicted by coral UV flu-orescence records, which are excellent record-ers of regional river discharge to the GBR and  provide an independent estimate of Queens-land’s precipitation ( 35 ). Fluorescence recordsindicate that river flow into the GBR continued throughout the last four centuries ( 14, 35 ). Inaddition, a multicentury drought should damp-en coral  18 O variability; instead, the amplitudeof interdecadal oscillations is greatest duringthe 18th century. Finally, the   18 O depletionafter    1870 is common to each of the south-western Pacific coral sites despite their differentrainfall regimes, proximities to terrestrial sourc-es, and land-use histories.Additionalprocessesthatinfluenceseawater   18 O uniformly throughout the southwesternPacific must be implicated in the 1870s centu-ry-scale  18 O shift. Both surface-ocean evapo-ration and advection are also important factorsmodifying SSS in the Coral Sea ( 36  ), and their contribution to the   18 O shift is supported by paleo-evidence for weakened trade winds and altered oceanic circulation since the mid-19thcentury. Globally, one of the most consistentfeaturesidentifiedinhistoricalandpaleoclimaterecords of the LIA is enhanced and highlyvariable atmospheric circulation ( 1–3, 5 ).Large-scalechangesintropicalatmosphericcir-culation are clearly evident in the PeruvianQuelccaya ice core [14°S, 71°W ( 6  )], which provides a high-resolution aeolian record of theSouthern Hemisphere trade wind belt (Fig. 3G).Greater dust deposition in the ice cap is indic-ative of increased wind velocities across thealtiplano of Southern Peru ( 6  ). During the LIAthe number of large aeolian particles (diameter   1.59  m) increased relative to the rest of theicecore,alongwithabrupt20%to30%shiftsinmicroparticles and conductivities ( 6  ). Signifi-cant correlations between the Quelccaya large particle concentrations and    18 O coral recordsfrom the southwestern Pacific ( r     0.57 for GBR coral   18 O) highlight the potential for intensified atmospheric circulation to affect both regions.We propose that during the LIA, strength-ened trade winds influenced the southwesternPacificevaporation-precipitationbalancebyen-hancing regional evaporation, causing the sea-water and coral to be enriched in  18 O. A changeof trade wind influence in the southwesternPacific is also evident in the varying strength of the western boundary current. The South Equa-torial Current (SEC) transports characteristical-ly high salinity (  35.6 psu) water westward into the Coral Sea before branching south alongthe GBR as the East Australian Current ( 37  ).Coral   14 C records from Abraham Reef ( 27  )and Heron Island, GBR ( 38 ) are consistent withthe scenario of a weakening of these currents as Fig. 3.  Comparison of region-al coral   18 O records andproxies of Southern Hemi-sphere ocean-atmospherecirculation. The compositeGBR   18 O record ( A ) is re-duced to a residual   18 Orecord ( B ) by subtracting thetemperature signal from theGBR Sr/Ca record. Propaga-tion of total uncertaintiesgives an average 95% confi-dence interval for (B) of   0.15‰ from 1700 to 1780and   0.11‰ from 1780 to1985. The three southwest-ern Pacific   18 O records are( C ) Abraham Reef, GBR(22°S, 153°E) ( 27 ), ( D )Amede´e Lighthouse, NewCaledonia (22°S, 166°E) ( 28 ),and ( E ) Espirtu Santos, Vanu-atu (15°S, 167°E) ( 29 ). Allrecords are resampled toequivalent 5-year averages,and each  18 O record is nor-malized to the average of thewhole record. ( F ) The annualAbraham Reef   14 C record of Druffel and Griffin ( 27 ) is cal-culated as an anomaly rela-tive to the expected oceanmixed-layer estimates of   14 C [values reported by ( 27 )in Fig. 1B]. The   14 C esti-mates were modeled from20-year tree-ring (atmo-spheric)   14 C values and anocean box-diffusion model( 39 ). ( G ) Number of largeparticles (diameter    1.59  m) in the Quelccaya icecore (14°S, 71°W) ( 6 ) plot-ted as an anomaly from the mean (1560 to 1980). The asterisk highlights the extreme dustconcentrations corresponding to the local eruption of Huaynaputina, Peru, in 1600 ( 6 ). R  E P O R T S www.sciencemag.org SCIENCE VOL 295 22 FEBRUARY 2002  1513  a result of lower wind stress since the LIA. Theabsence of the “Suess effect” (the lowering of   14 C values from the late 19th century as fossilfuel burning released   14 C-free CO 2 ) in theserecords is interpreted as a long-term change inocean circulation within the southwestern Pa-cific since the late 1800s ( 38 ). Rising   14 Cvalues since the 1880s indicate a greater influxof younger surface waters, and this reduced oceanic “reservoir effect” overrides the antici- pated Suess effect (Fig. 3F). We interpret thisshift as a weakening of the wind-driven SECsince the late 19th century, which reduced thetransport of upwelled equatorial  14 C-depleted water to the GBR corals.Our SSTA reconstructions suggest that astronger latitudinal temperature gradient, rel-ative to the present, may have prevailed dur-ing the LIA. General circulation model ex- periments ( 9 ) indicate that the global climateresponds to an exaggerated latitudinal tem- perature difference by intensifying the large-scale atmospheric dynamics of the Hadleycirculation cells in order to maintain thermal balance. During the LIA, if warm tropicalPacific SSTs encouraged evaporation (as im- plied by the coral  18 O records) and a stron-ger Hadley circulation dried the subtropics,then global average water vapor should haveincreased. A critical question is where thatmoisture was transported. Generally, snowcover increases with a strong latitudinal tem- perature gradient, in part because of cooler high-latitude conditions, but also as a resultof a greater atmospheric connection betweenthe tropics and extratropics ( 10 ). Our resultsimply that the tropical oceans may have played an important role in driving the LIAglacial expansion during the repeated advanc-es between 1600 and 1860 ( 1 ). Cooling and abrupt freshening of the tropical southwest-ern Pacific coincided with the weakening of atmospheric circulation at the end of the LIA,when glaciers worldwide began to retreat. References and Notes 1. J. M. Grove,  The Little Ice Age  (Methuen, London,1988).2. T. J. Crowley, G. R. North,  Paleoclimatology   (OxfordUniv. Press, New York, 1991).3. H. H. Lamb,  Climate, History and the Modern World  (Routledge, London, ed. 2, 1995).4. R. S. Bradley, P. D. Jones,  The Holocene  3 , 367 (1993).5. K. J. Kreutz  et al .,  Science  277 , 1294 (1997).6. L. G. Thompson, E. Mosley-Thompson, W. Dansgaard,P. M. Grootes,  Science  234 , 361 (1986).7. P. D. Jones, K. R. Briffa, T. P. Barnett, S. F. B. Tett,  TheHolocene  8 , 455 (1998).8. P. D. Jones, T. J. Osborn, K. R. Briffa,  Science  292 , 662(2001).9. D. Rind,  J. Geophys. Res.  103 , 5943 (1998).10.    ,  Quat. Sci. Rev.  19 , 381 (2000).11. See supplemental material on  Science  Online atwww.sciencemag.org/cgi/content/full/295/5559/1511/DC1.12. M. T. McCulloch, M. K. Gagan, G. E. Mortimer, A. R.Chivas, P. J. Isdale,  Geochim. Cosmochim. Acta  58 ,2747 (1994).13. M. K. Gagan  et al .,  Science  279 , 1014 (1998).14. E. J. Hendy, M. K. Gagan, J. M. Lough, in preparation.15. Sr/Ca and U/Ca were determined by isotope dilutionand analyzed by thermal ionization mass spectrom-etry and solution inductively coupled plasma massspectrometry, respectively. For methodological de-tails, see ( 11 ).16. The composite reconstructions are the average of allrecords, after normalization relative to the longestcontinuous record (HAV-01B). The  18 O composite isbased on eight cores; the Sr/Ca and U/Ca compositesare constructed from seven (PAN08B excluded). Error bounds were calculated using 95% confidence inter-vals for each 5-year period. Periods of higher U/Cavariability coincide with die-offs in two of the coresin 1782–1785 and 1817. Data are available at www.ngdr.noaa.gov/paleo/pubs/hendy2002.17. C. Alibert, M. T. McCulloch,  Paleoceanography   12 ,345 (1997).18. G. R. Min  et al .,  Geochim. Cosmochim. Acta  59 , 2025(1995).19. N. A. Rayner, E. B. Horton, D. E. Parker, C. K. Folland,K. B. Hackett,  Version 2.2 of the Global Sea–Ice and  Sea Surface Temperature Data Set, 1903–1994  (Had-ley Centre for Climate Prediction and Research, Me-teorological Office, Bracknell, Berkshire, UK, 1996).20. M. Bottomley, C. K. Folland, J. Hsiung, R. E. Newell,D. E. Parker,  Global Ocean Surface Temperature Atlas “ GOSTA ” (U.K. Meterological Office, Her Majesty’sStationary Office, London, 1990).21. B. K. Linsley, G. M. Wellington, D. P. Schrag,  Science 290 , 1145 (2000).22. K. R. Briffa, P. D. Jones, F. H. Schweingruber, T. J.Osborn,  Nature  393 , 450 (1998).23. M. E. Mann, R. S. Bradley, M. K. Hughes,  Nature  392 ,779 (1998).24. T. J. Crowley,  Science  289 , 270 (2000).25. P. K. Swart, M. L. Coleman,  Nature  283 , 557 (1980).26. The  18 O residual is calculated as  18 O   18 O/  T  ( T   18O  T  Sr/Ca )where the temperature-dependent function,   18 O/  T  , is –0.18/°C for   Porites  sp. ( 13 ).27. E. R. M. Druffel, S. Griffin,  J. Geophys. Res.  98 , 20(1993).28. T. M. Quinn  et al .,  Paleoceanography   13 , 412 (1998).29. T. M. Quinn, F. W. Taylor, T. J. Crowley,  Quat. Sci.Rev.  12 , 407 (1993).30. M. N. Evans, A. Kaplan, M. A. Cane,  Paleoceanography  15 , 551 (2000).31. T. J. Crowley, T. M. Quinn, W. T. Hyde,  Paleoceanog-raphy   14 , 605 (1999).32. B. K. Linsley, R. B. Dunbar, G. M. Wellington, D. A.Mucciarone,  J. Geophys. Res.  99 , 9977 (1994).33. E. Wolanski,  Physical Oceanographic Processes of theGreat Barrier Reef   (CRC Press, Boca Raton, FL, 1994).34. J. M. Lough,  Int. J. Clim.  17 , 55 (1997).35. P. J. Isdale, B. J. Stewart, K. S. Tickle, J. M. Lough,  TheHolocene  8 , 1 (1998).36. T. Delcroix, C. Henin, V. Porte, P. Arkin,  Deep-Sea Res. 43 , 1123 (1996).37. S. Sokolov, S. Rintoul,  J. Mar. Res.  58 , 223 (2000).38. E. R. M. Druffel, S. Griffin,  J. Geophys. Res.  104 , 23607(1999).39. M. Stuiver, G. W. Pearson, T. F. Braziunas,  Radiocar-bon  28 , 980 (1986).40. We thank L. Kinsley, H. Scott-Gagan, and J. Cali for analytical assistance; S. Fallon and G. Mortimer for the U/Ca method development; B. Parker and M.Devereux for assistance with the corals; and J. Chap-pell, C. Hendy, G. Meyers, and two anonymous re-viewers for valuable comments on the manuscript.E.J.H. was supported by an Australian PostgraduateAward.2 November 2001; accepted 22 January 2002 Microbial Activity at GigapascalPressures Anurag Sharma,* James H. Scott,* George D. Cody,Marilyn L. Fogel, Robert M. Hazen, Russell J. Hemley,Wesley T. Huntress We observed physiological and metabolic activity of   Shewanella oneidensis strain MR1 and  Escherichia coli   strain MG1655 at pressures of 68 to 1680megapascals (MPa) in diamond anvil cells. We measured biological formateoxidationathighpressures(68to1060MPa).Atpressuresof1200to1600MPa,living bacteria resided in fluid inclusions in ice-VI crystals and continued to beviable upon subsequent release to ambient pressures (0.1 MPa). Evidence of microbial viability and activity at these extreme pressures expands by an order of magnitude the range of conditions representing the habitable zone in thesolar system. Microbialcommunitiesadapttoawiderangeof  pressures, temperatures, salinities, pH, and ox-idation states. Although the chemical and phys-ical conditions in these extreme environmentsare reasonably well constrained, the conse-quence of these physical parameters on the physiology of microbial communities is notwell understood. Significant attention has beenfocused on the effects of high and low temper-ature on physiology ( 1, 2 ). There is some evi-dence that elevated pressure may also manifestinteresting effects on cellular physiology ( 3, 4 ).For example, recent studies report that elevated  pressure may lead to enzyme inactivation, com- promise cell-membrane integrity, and suppress protein interactions with various substrates ( 3  –  6  ). Whereas the cumulative impact of these pressure-induced effects on microbial metabo-lism and physiology is an inhibition in growthrate and cellular division in microorganisms( 7   –  9 ), exactly how these factors affect intactcells is not well understood ( 4, 10 ). Numerous high-pressure studies have beenconducted on biological systems; however,these have been either on individual biomol- Geophysical Laboratory, Carnegie Institution of Washington, 5251 Broad Branch Road, N.W., Wash-ington, DC 20015, USA.*To whom correspondence should be addressed. E-mail: sharma@gl.ciw.edu (A.S.); j.scott@gl.ciw.edu(J.H.S.) R  E P O R T S 22 FEBRUARY 2002 VOL 295 SCIENCE www.sciencemag.org 1514
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks